Research map / Bandits and online learning / Multi-armed bandits
Bayesian Design Principles for Frequentist Sequential Learning
Yunbei Xu, assaf zeevi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use Adaptive Posterior Sampling (APS) instead of UCB 1 | 16-armed stochastic Bernoulli bandit · Stochastic Bernoulli MAB | expected regret | Improved |
| Use APS instead of EXP3 | 16-armed adversarial bandit · Adversarial Bernoulli MAB | expected regret | Improved |
| Use APS instead of EXP3 | 16-armed non-stationary bandit with change points · Non-stationary Bernoulli MAB (change points) | dynamic regret | Improved |
| Use APS instead of restarted EXP3 or restarted UCB 1 | 16-armed non-stationary bandit with change points · Non-stationary Bernoulli MAB (change points) | dynamic regret | Improved |
| Use APS instead of restarted Thompson Sampling | 16-armed non-stationary bandit with change points · Non-stationary Bernoulli MAB (change points) | dynamic regret | No clear effect |
| Use APS instead of Thompson Sampling | 4-armed non-stationary bandit with sine-curve rewards · Non-stationary Bernoulli MAB (sine curves) | regret | Improved |
| Create a fresh algorithmic belief each round rather than relying on a fixed prior | APS · Non-stationary Bernoulli MAB (sine curves) | responsiveness to the best arm | Improved |
Closest papers by meaning
- Meta-Learning for Simple Regret Minimization (AAAI 2023 · similarity 0.90)
- Bayesian Regret Minimization in Offline Bandits (ICML 2024 · similarity 0.89)
- Only Pay for What Is Uncertain: Variance-Adaptive Thompson Sampling (ICLR 2024 · similarity 0.89)
- Towards Optimal Regret in Adversarial Linear MDPs with Bandit Feedback (ICLR 2024 · similarity 0.88)
- Noise-Adaptive Confidence Sets for Linear Bandits and Application to Bayesian Optimization (ICML 2024 · similarity 0.88)
- Adaptive Regret for Bandits Made Possible: Two Queries Suffice (ICLR 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
